The degree of tendon retraction plays a role in determining surgical reparability. The tendon is less likely to withstand the tensile load imparted on the repaired tendon when there is a greater degree of tendon retraction. This may tend toward a poor prognosis, and it is therefore important to assess at the time of MR imaging. As medial retraction of the tendon approaches the glenoid fossa, the more likely the tendon will be irreparable [12] (Fig. 3.1 and see Pearls and Pitfalls), though be aware that the medial extent of tendon retraction as demonstrated at MR imaging may not represent true tendon retraction and may be mobile at arthroscopy. This typically depends on the chronicity of the tear as acute tears are more mobile than those that are chronic.

Although tendon tear shape is evaluated arthroscopically, the ability to detect tendon tear shapes decreases as the size of the tendon tear increases [13] with a limited role for MR imaging and less so for ultrasound. Massive contracted rotator cuff tears have been categorized into massive crescentic tears (wide anteroposterior dimension) and massive longitudinal tears (spared anterior cuff tissue at the rotator interval) [14].

Many cuff tears are reparable [14], and the ability to detect an acute traumatic massive cuff tear or avulsion rather than a chronic tear is important, as the acute tear will have good-quality elastic tendon stump and will not place undue tension on the tissue following repair. MR imaging can provide useful information about acuity and tendon quality given the degree or absence of soft tissue edema signal (Fig. 3.3). When the tear is chronic, it is important to evaluate the presence of degeneration characterized by fraying and irregularity and possible scarring, as the presumably poor-quality and/or inelastic tendon is unlikely to withstand suture anchoring and tension resulting in a poor postoperative prognosis (see Pearls and Pitfalls) [8].

Although degree of tendon retraction and tendon quality is important, it is the integrity of muscle quality that is integral in formulating a treatment plan. Fatty infiltration of the rotator cuff muscle belly has been established as a negative prognostic factor for reparability of the rotator cuff [12, 15–17]. It is therefore important to accurately assess the muscle quality using imaging, as surgical management and prognosis will be influenced by the status of muscle quality. The Goutallier classification of fatty infiltration, initially described using CT, is often applied in practice with MRI and US as they are currently the imaging modalities of choice in assessing the rotator cuff tendons. In clinical practice, modification of the Goutallier classification to a three-tiered staging system is used to indicate the presence of fatty infiltration (Fig. 3.4): mild fatty infiltration (Goutallier stage 0 and stage 1) indicates normal muscle without or with minimal fatty streaks; moderate fatty infiltration (Goutallier stage 2) is characterized by fatty infiltration, but more muscle than fat; and severe fatty infiltration (Goutallier stage 3 and stage 4) demonstrates at least as much fatty infiltration as there is muscle. Warner et al. evaluated the degree of fatty infiltration in massive rotator cuff tears of the same size and found correlation between the degree of fatty infiltration on MRI and overall biomechanics and function [11, 15].

Ultrasound can be used to evaluate the degree of muscle atrophy and fatty infiltration with good correlation with MR imaging [18] – nevertheless, this remains challenging and is user dependent. Fatty infiltration will be seen as effacement of the normal pennate pattern of the muscle and increased hyperechogenicity (Fig. 3.5). It is often helpful to compare the degree of echogenicity with the overlying deltoid or trapezius muscles and the adjacent teres minor muscle when they are normal.

Methods to assess supraspinatus muscle atrophy include the scapular occupation ratio and the “tangent sign.” It is important to recognize the difference between loss of muscle bulk and true fatty infiltration. Muscle atrophy has the potential to be reversible if fatty infiltration is not present. Zanetti and coworkers described the “tangent sign” to assess muscle atrophy in the sagittal oblique plane on MRI at the medial coracoid process [19] (Fig. 3.6). The normal supraspinatus muscle belly should cross superior to a line across the superior borders of the scapular spine and the superior margin of the coracoid process. With an atrophic muscle belly, the muscle will lie inferior to the line.

Thomazeau described the occupation ratio, which is evaluated in the same imaging plane and utilizing the same osseous landmarks [12]. It is defined as the ratio of the cross-sectional area of the supraspinatus muscle to the area of the supraspinatus fossa, with a ratio of less than 50 % indicative of muscle atrophy (Fig. 3.7). This method was shown to correlate supraspinatus atrophy with the extent of tendon tear and was associated with postoperative tear recurrence. Both methods have been shown to correlate with strength and mobility. When evaluating the cross-sectional area of the supraspinatus and other cuff tendons, it is important to be aware of the degree of tendon retraction as this can result in an overestimation of muscle atrophy if the torn tendon has retracted far medially (see Pearls and Pitfalls). Cross-referencing with the coronal and axial planes and awareness of the marbling due to fatty infiltration should help one avoid this imaging pitfall (Fig. 3.8). Cross-sectional areas of the infraspinatus, teres minor, and subscapularis muscles have also been correlated with chronicity of rotator cuff tears, with the subscapularis muscle atrophy being associated with poorer surgical outcomes [11, 19].

It is important to recognize subacute muscle denervation involving one or more muscle groups of the shoulder girdle, such as seen with Parsonage-Turner syndrome (inflammation of the brachial plexus). Although, subacute denervation is rare, it can mimic the clinical presentation of pseudoparalysis seen with a massive rotator cuff tear (see Pearls and Pitfalls). Subacute muscle denervation causes diffuse homogenous edema signal confined to the involved muscle and is usually evident on MR imaging 2–4 weeks after denervation has occurred [20] (Fig. 3.9). This can be distinguished from acute muscle injury, as muscle injury characteristically manifests as a feathery edema pattern and often has adjacent edema signal within the soft tissues. Acute, traumatic edema is typically seen within hours to days after injury. If innervation is restored, the MR imaging findings will return to normal. If not restored, a chronic irreversible denervation will ensue characterized by fatty infiltration. If denervation is suspected, correlation with electromyographic findings and a careful search for upper extremity peripheral nerve compression by a mass lesion should be performed. Parsonage-Turner syndrome is usually self-limited with a return to full recovery as the rule.